Non-autoregressive Transformer(NAT) significantly accelerates the inference of neural machine translation. However, conventional NAT models suffer from limited expression power and performance degradation compared to ...
详细信息
A hybrid subgrid scheme based on the conventional finite-difference time-domain (FDTD) schemes are proposed. The alternating-direction-implicit FDTD (ADI-FDTD) is used to calculate electromagnetic fields in fine grid ...
详细信息
IOTA (Internet of Things Applications) protocol aims to provide a secure and trustworthy protocol for realizing data-value interactions between IoT devices, and it has become the next possible direction for the develo...
IOTA (Internet of Things Applications) protocol aims to provide a secure and trustworthy protocol for realizing data-value interactions between IoT devices, and it has become the next possible direction for the development of Blockchain applications. IOTA protocol has the problem that the time-consuming cumulative weight update grows too fast with the increase of the number of transactions, this paper proposed an improved IOTA protocol based on optimal midway walk, the improved protocol uses cutting-edge transaction selection algorithm of weighted Monte Carlo Markov random walk to reduce cumulative weight update delay. Firstly, this improved IOTA protocol mainly ensures the efficiency of transaction confirmation without losing the security of consensus by introducing the optimal midway walk method, i.e., the starting point of each walk is chosen to depart at an optimal location. Secondly, this paper designs the indirect verification identity record algorithm and the layer degree of confirmation principle are designed to realize the improved IOTA protocol's resilience and resistance to Parasite chain attack. The experimental results show that the improved IOTA using random function to select the midway walk starting point from i-50λ, to i-75λsubstantially solves the system computational burden problem and greatly reduces the cumulative weight update delay.
Ahstract-A new low-profile quadrifilar helical antenna (QHA) is proposed and investigated. Four spiral arms of the QHA are divided and connected to for a sleeve structure to reduce the height of the antenna. In this c...
详细信息
Ahstract-A new low-profile quadrifilar helical antenna (QHA) is proposed and investigated. Four spiral arms of the QHA are divided and connected to for a sleeve structure to reduce the height of the antenna. In this case, four spiral arms on the inner cylindrical sleeve are connected with the ground, while the upper ends of the four spiral arms on the outer cylindrical sleeve are short-circuited, and inner and outer spiral arms are connected in the middle. The QHA has a beam width reaches to above $120 ^{\circ}$ and a cardioid pattern at 1.575 GHz. As a result, the QHA has a very low profile of only 22 mm. The simulation and optimization from the CST show that the center frequency of the QHA is 1.575 GHz, which is used for the satellite navigation system, with axis ratio in the operating band less than 3 dB.
This paper introduces the application of unconditionally stable locally one-dimensional finite-difference time-domain(LOD-FDTD) method to 3D multi-pole Debye dispersive media model. Compared with other methods for dis...
详细信息
A sparse array (SA), termed as ATLF3-2BL, is proposed based on the uniform linear array (ULA) fitting (UF) principle. UF is a general SA devising scheme, based on which designers can devise SAs using a series of ULAs....
详细信息
A sparse array (SA), termed as ATLF3-2BL, is proposed based on the uniform linear array (ULA) fitting (UF) principle. UF is a general SA devising scheme, based on which designers can devise SAs using a series of ULAs. SAs devised upon UF enjoy closed-form expressions for both array configuration and uniform degrees of freedom. Comprised of 7 ULAs, the proposed ATLF3-2BL can provide a long virtual difference coarray than most existing SAs. Besides, the mutual coupling in ATLF3-2BL remains a low level. Therefore, the ATLF3-2BL gives a considerable compromise between mutual coupling and uniform degrees of freedom. Simulations attest the good performance of the proposed ATLF3-2BL.
A high-order spatial filtering-symplectic finite difference time domain (SF-SFDTD) scheme with controllable stability condition is proposed to solve the time-dependent Schrödinger equation (TDSE). Firstly, the hi...
详细信息
A General Constrained Adaptive Filtering (GCAF) algorithm is proposed via constructing a general and adaptive loss function to find out a solution of constraint optimizing problem. By selecting appropriate parameters,...
详细信息
This paper presents a hybrid method for determination of continuous dielectric properties of clothing materials. The dielectric constant and loss tangent of three types of materials are firstly investigated using open...
This paper presents a hybrid method for determination of continuous dielectric properties of clothing materials. The dielectric constant and loss tangent of three types of materials are firstly investigated using open-ended coaxial probe from 0.5 to 20 GHz. The average measured values are input into the CST simulation of two CPW lines to obtain the two-port S-parameters. Finally, more reasonable yet reliable results are obtained by using two retrieval algorithms (TRACE and ABCD) for validation and comparative purposes.
Patent classification is an essential task in patent information management and knowledge mining. Most existing studies are based on the textual content of individual patent texts (e.g., titles and abstracts) for clas...
详细信息
ISBN:
(纸本)9781665416672
Patent classification is an essential task in patent information management and knowledge mining. Most existing studies are based on the textual content of individual patent texts (e.g., titles and abstracts) for classification, but the patent also has label information with hierarchical structure and semantic description. However, the semantics correlation of patent texts and their labels have been largely ignored. In this paper, we propose a novel framework called HLSPC for patent classification by leveraging the hierarchical semantics correlation of patent texts and their labels. Specifically, we first apply a patent representation learning module for capturing the semantics representation of patent texts and hierarchical labels. Then, we design a label attention learning module to build the semantics correlation between patent texts and hierarchical labels, which enhances patent representation. Finally, we deploy a multi-level fusion module to get the refined category prediction for each patent which can preserve both local and global hierarchical prediction information. Extensive experimental results on two English patent datasets demonstrate the effective power of the HLSPC model.
暂无评论